Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
173 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Unbiased approximation of the ergodic measure for piecewise $α$-stable Ornstein-Uhlenbeck processes arising in queueing networks (2405.18851v2)

Published 29 May 2024 in math.PR

Abstract: Piecewise $\alpha$-stable Ornstein-Uhlenbeck (OU) processes arising in queue networks usually do not have an explicit dissipation, which makes the related numerical methods such as Euler-Maruyama (EM) scheme more difficult to analyze. We develop an EM scheme with decreasing step size $\Lambda=(\eta_n)_{n\in \mathbb{N}}$ to approximate their ergodic measures. This approximation does not have a bias and has a rate $\eta{1/\alpha}_n$ in Wasserstein-1 distance. We show by the classical OU process that our convergence rate is optimal. We further prove the central limit theorem (CLT) and moderate derivation principle (MDP) for the empirical measure of these piecewise $\alpha$-stable Ornstein-Uhlenbeck processes. In addition, we use the Sinkhorn--Knopp algorithm to compute the Wasserstein-1 distance and conduct simulations for several concrete examples.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (51)
  1. The limit of stationary distributions of many-server queues in the Halfin-Whitt regime. Mathematics of Operations Research 45, 3 (2020), 1016–1055.
  2. On uniform exponential ergodicity of Markovian multiclass many-server queues in the Halfin-Whitt regime. Mathematics of Operations Research 46, 2 (2021), 772–796.
  3. Uniform polynomial rates of convergence for a class of Lévy-driven controlled SDEs arising in multiclass many-server queues. In modeling, stochastic control, optimization, and applications, vol. 164 of IMA Vol. Math. Appl. Springer, Cham, 2019, pp. 1–20.
  4. Ergodicity of a Lévy-driven SDE arising from multiclass many-server queues. The Annals of Applied Probability 29, 2 (2019), 1070–1126.
  5. Systems of equations driven by stable processes. Probability Theory and Related Fields 134, 2 (2006), 175–214.
  6. Ergodicity of inhomogeneous Markov chains through asymptotic pseudotrajectories. The Annals of Applied Probability 27, 5 (2017), 3004–3049.
  7. Nonnegative matrices in the Mathematical sciences, vol. 9 of Classics in Applied Mathematics. Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 1994. Revised reprint of the 1979 original.
  8. Introduction to linear optimization, vol. 6 of Athena Scientific books. Athena Scientific Belmont, MA, 1997.
  9. Stein’s method for steady-state diffusion approximations of M/P⁢h/n+M𝑀𝑃ℎ𝑛𝑀M/Ph/n+Mitalic_M / italic_P italic_h / italic_n + italic_M systems. The Annals of Applied Probability 27, 1 (2017), 550–581.
  10. Stein’s method for steady-state diffusion approximations: An introduction through the Erlang-A and Erlang-C models. Stochastic Systems 6, 2 (2016), 301–366.
  11. Piecewise-linear diffusion processes. In Advances in Queueing Theory, Methods, and Open Problems. CRC Press, 1995, pp. 463–480.
  12. A numerical scheme for invariant distributions of constrained diffusions. Mathematics of Operations Research 39, 2 (2014), 262–289.
  13. Approximation of the invariant measure of stable SDEs by an Euler-Maruyama scheme. Stochastic Processes and their Applications 163 (2023), 136–167.
  14. Approximation of the invariant measure for stable SDE by the Euler-Maruyama scheme with decreasing step-sizes, 2023.
  15. Solving inverse stochastic problems from discrete particle observations using the Fokker–Planck equation and Physics-Informed Neural networks. SIAM Journal on Scientific Computing 43, 3 (2021), B811–B830.
  16. Faster Wasserstein distance estimation with the Sinkhorn divergence. In Advances in Neural Information Processing Systems (2020), H. Larochelle, M. Ranzato, R. Hadsell, M. Balcan, and H. Lin, Eds., vol. 33, Curran Associates, Inc., pp. 2257–2269.
  17. Cuturi, M. Sinkhorn distances: Lightspeed computation of optimal transport. In Advances in Neural Information Processing Systems (2013), C. Burges, L. Bottou, M. Welling, Z. Ghahramani, and K. Weinberger, Eds., vol. 26, Curran Associates, Inc.
  18. Many-server queues with customer abandonment: Numerical analysis of their diffusion model. Stochastic Systems 3, 1 (2013), 96–146.
  19. Many-server diffusion limits for G/P⁢h/n+G⁢I𝐺𝑃ℎ𝑛𝐺𝐼G/Ph/n+GIitalic_G / italic_P italic_h / italic_n + italic_G italic_I queues. Annals of Applied Probability 20, 5 (2010), 1854–1890.
  20. Numerical analysis for diffusion models of G⁢I/P⁢h/n+G⁢I𝐺𝐼𝑃ℎ𝑛𝐺𝐼GI/Ph/n+GIitalic_G italic_I / italic_P italic_h / italic_n + italic_G italic_I queues. ACM SIGMETRICS Performance Evaluation Review 39, 4 (2012), 37–37.
  21. Wasserstein-1111 distance between SDEs driven by Brownian motion and stable processes. arXiv preprint arXiv:2302.03372 (2023).
  22. Positive recurrence of piecewise Ornstein-Uhlenbeck processes and common quadratic Lyapunov functions. The Annals of Applied Probability 23, 4 (2013), 1291–1317.
  23. Multivariate approximations in Wasserstein distance by Stein’s method and Bismut’s formula. Probability Theory and Related Fields 174, 3-4 (2019), 945–979.
  24. Flury, B. D. Acceptance-rejection sampling made easy. SIAM Review 32, 3 (1990), 474–476.
  25. Gurvich, I. Diffusion models and steady-state approximations for exponentially ergodic Markovian queues. The Annals of Applied Probability 24, 6 (2014), 2527–2559.
  26. Hairer, M. Convergence of Markov processes. Lecture notes 18 (2010), 26.
  27. Topics in matrix analysis. Cambridge University Press, Cambridge, 1994. Corrected reprint of the 1991 original.
  28. The Euler-Maruyama method for S(F)DEs with Hölder drift and α𝛼\alphaitalic_α-stable noise. Stochastic Analysis and Applications 36, 1 (2018), 28–39.
  29. An approximation to the invariant measure of the limiting diffusion of G/P⁢h/n+G⁢I𝐺𝑃ℎ𝑛𝐺𝐼{G}/{P}h/n+{GI}italic_G / italic_P italic_h / italic_n + italic_G italic_I queues in the Halfin–Whitt regime and related asymptotics. Mathematics of Operations Research (2024).
  30. Optimal transport-inspired deep learning framework for slow-decaying problems: Exploiting Sinkhorn loss and Wasserstein kernel, 2023.
  31. Klebaner, F. C. Introduction to stochastic calculus with applications, third ed. Imperial College Press, London, 2012.
  32. Knight, P. A. The Sinkhorn-Knopp algorithm: Convergence and applications. SIAM Journal on Matrix Analysis and Applications 30, 1 (2008), 261–275.
  33. Strong convergence of the Euler-Maruyama approximation for a class of Lévy-driven SDEs. Stochastic Processes and their Applications 129, 8 (2019), 2654–2680.
  34. Fast Sinkhorn I: An O(N) algorithm for the Wasserstein-1 metric. arXiv preprint arXiv:2202.10042 (2022).
  35. Differential properties of Sinkhorn approximation for learning with Wasserstein distance. In Advances in Neural Information Processing Systems (2018), S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett, Eds., vol. 31, Curran Associates, Inc.
  36. Strong rate of convergence for the Euler-Maruyama approximation of SDEs with Hölder continuous drift coefficient. Stochastic Processes and their Applications 127, 8 (2017), 2542–2559.
  37. Nolan, J. P. Univariate stable distributions: Models for heavy tailed data. Springer Series in Operations Research and Financial Engineering. Springer, Cham, 2020.
  38. Unadjusted Langevin algorithm with multiplicative noise: Total variation and Wasserstein bounds. The Annals of Applied Probability, 1 (2023), 726 – 779.
  39. Martingale proofs of many-server heavy-traffic limits for Markovian queues. Probability Surveys 4 (2007), 193–267.
  40. Continuity of a queueing integral representation in the M1 topology. The Annals of Applied Probability 20, 1 (2010), 214 – 237.
  41. Panloup, F. Computation of the invariant measure for a Lévy driven SDE: Rate of convergence. Stochastic Processes and their Applications 118, 8 (2008), 1351–1384.
  42. Panloup, F. Recursive computation of the invariant measure of a stochastic differential equation driven by a Lévy process. The Annals of Applied Probability 18, 2 (2008), 379–426.
  43. The Euler scheme for Lévy driven stochastic differential equations. The Annals of Probability 25, 1 (1997), 393–423.
  44. The multiclass G⁢I/P⁢H/N𝐺𝐼𝑃𝐻𝑁GI/PH/Nitalic_G italic_I / italic_P italic_H / italic_N queue in the Halfin-Whitt regime. Advances in Applied Probability 32, 2 (2000), 564–595.
  45. Algorithmic stability of heavy-tailed SGD with general loss functions. In International Conference on Machine Learning (2023), PMLR, pp. 28578–28597.
  46. Stable non-Gaussian random processes. Stochastic Modeling. Chapman & Hall, New York, 1994. Stochastic models with infinite variance.
  47. Tankov, P. Financial modelling with jump processes. Chapman and Hall/CRC, 2003.
  48. Villani, C. Optimal transport, vol. 338 of Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences]. Springer-Verlag, Berlin, 2009. Old and new.
  49. Wang, J. Exponential ergodicity and strong ergodicity for SDEs driven by symmetric α𝛼\alphaitalic_α-stable processes. Applied Mathematics Letters. An International Journal of Rapid Publication 26, 6 (2013), 654–658.
  50. Wang, J. Lpsuperscript𝐿𝑝L^{p}italic_L start_POSTSUPERSCRIPT italic_p end_POSTSUPERSCRIPT-Wasserstein distance for stochastic differential equations driven by Lévy processes. Bernoulli 22, 3 (2016), 1598–1616.
  51. A fast proximal point method for computing exact Wasserstein distance. In Proceedings of The 35th Uncertainty in Artificial Intelligence Conference (22–25 Jul 2020), R. P. Adams and V. Gogate, Eds., vol. 115 of Proceedings of Machine Learning Research, PMLR, pp. 433–453.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now